Pith. sign in

REVIEW 3 major objections 7 minor 1 cited by

A framework for realisable data-driven active flow control using model predictive control applied to a simplified truck wake

T0 review · 3 major / 7 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read An offline-trained latent MPC cuts truck-wake drag by 12.8%

desk verdict A sensible, clearly-written integration of known pieces that gives a plausible 12.8% drag reduction in a 2D DNS, but the headline number rests on one run with no simple forcing baseline; worth refereeing, not yet a robust result. read the letter →

arxiv 2510.11600 v3 pith:QHUBWMBU submitted 2025-10-13 physics.flu-dyn

Alberto Solera-Rico , Carlos Sanmiguel Vila , Stefano Discetti This is my paper · ORCID
classification physics.flu-dyn
keywords activeflowcontrolmodelpredictiveLSTMlatentdynamicssensorselectionSHAPdragreductiontruckwake
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that a complete closed-loop active flow control system can be assembled entirely from offline data, using only a handful of non-intrusive surface pressure sensors. In two-dimensional DNS of a simplified truck wake at Reynolds number 500, the resulting controller—built from an LSTM encoder, a latent dynamics model, and model predictive control—reduces mean drag by 12.8% and raises base pressure by more than 35%. The authors' central claim is that the wake state can be inferred from surface pressure history, propagated through a low-dimensional latent space, and optimised by cheap gradient-based MPC, with no online learning. If correct, this offers a practical path from demonstrations to deployable real-time control on road vehicles.

What carries the argument

The central object is the differentiable latent dynamics chain: an LSTM encoder (state estimator), a residual MLP that advances the latent state under a control action, and a force decoder. Because the entire chain supports automatic differentiation, the model predictive controller's cost—mean drag, drag fluctuation, lift magnitude, and control-signal smoothness—can be minimised by exact gradient backpropagation over a 25-step horizon, making the optimisation cheap enough for receding-horizon real-time use. The latent space is regularised to be decorrelated and force-informative, and it develops a V-shaped structure whose arms correspond to opposite signs of lift, reflecting the vortex-shedd

What would settle it

Repeat the identical closed-loop experiment at a different Reynolds number (for instance, Re=1500) or with a perturbed control path, and measure the drag reduction; if the latent model's predictions diverge once the wake is stabilised, the reported gain is tied to the single run rather than to the framework.

Watch

Extended reading notes

Core claim

The paper demonstrates the complete workflow end to end. A 90-sensor pressure array records the wake's pressure footprint during randomly modulated open-loop jet actuation; an LSTM-based temporal encoder maps 32 steps of this history into an 8-dimensional latent state; an MLP dynamics model predicts the latent-state increment under the current jet action; and a decoder maps the latent state to drag and lift. After training, a Shapley-value attribution analysis ranks the sensors and a slim encoder is distilled to operate on only four base-located probes. In closed-loop DNS, the model predictive controller reduces mean drag coefficient from 1.051 to 0.916, raises mean base pressure from -0.441

Load-bearing premise

The 12.8% reduction rests on the trained latent-dynamics model staying accurate while the controller drives the wake away from the data it was trained on—the paper notes the model 'departs further from its interpolation region' as the wake stabilises, and a single simulation run does not quantify that risk.

Editorial extensions

If this is right

  • A controller trained entirely on open-loop actuator data can be deployed closed-loop without online learning, and in this test case it reduces drag by 12.8%.
  • Four surface pressure probes located near the base are sufficient to reconstruct the wake state for control, eliminating the need for intrusive wake probes.
  • Because the MPC cost function is explicit, the same trained model could be reused to balance drag reduction against energy consumption or actuator wear at deployment time.
  • The observed mechanism is wake stabilisation: a longer recirculation bubble, weaker base suction, and reduced transverse velocity fluctuations (standard deviation down more than 23%).

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural next step is to test robustness: adding measurement noise, gusts, or a Reynolds-number change to the DNS would probe whether the 12.8% figure survives conditions the model was not trained on.
  • The Shapley-based sensor ranking is never benchmarked against an exhaustive search over four-sensor subsets; a directed comparison would bound how close the selected sensors are to the best possible placement.
  • The eight-dimensional latent space is hypothesised to hold for a 2D wake at Re=500; whether the same encoder-dynamics-decoder architecture transfers to three-dimensional bluff-body wakes is untested.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. The paper presents an end-to-end data-driven active flow control framework that combines an offline-trained LSTM-based latent dynamics model with gradient-based model predictive control (MPC). The latent state is inferred from surface-mounted pressure sensors, and SHAP-based attribution is used to reduce the sensor set from 90 to four. The method is demonstrated in a 2D DNS of a simplified truck wake at Re=500, where the authors report a 12.8% drag reduction (Cd from 1.051 to 0.916) under closed-loop MPC. The drag reduction is directly measured in the DNS and the methodology is described in detail. However, the central demonstration rests on a single closed-loop trajectory, with no repeated runs, no uncertainty quantification, and no comparison against simpler open-loop forcing. The paper also acknowledges that the model is extrapolating in the closed-loop regime (§3.3), which raises a specific concern about the reliability of the MPC-optimized actions.

Significance. If the reported drag reduction is robust, this is a valuable contribution to data-driven flow control. The framework addresses practical deployment constraints—offline training, few non-intrusive surface sensors, and a differentiable latent model that makes MPC computationally tractable. The SHAP-based sensor selection is interpretable and the four-sensor ``slim'' encoder achieves force-prediction accuracy comparable to the full 90-sensor model on the test set (§3.2, Fig. 10). The 12.8% drag reduction is a direct DNS measurement, not an artifact of the model. The main limitation is the lack of statistical and comparative evidence: the headline result is a single run, and the paper does not establish that the MPC/latent-model machinery adds value over simpler periodic forcing or that the result is repeatable across initial conditions.

major comments (3)
  1. [§3.3, Fig. 11; Eq. (13)] The 12.8% drag reduction is computed from a single closed-loop DNS run over 100 convective times. Since the uncontrolled wake at Re=500 is a limit-cycle flow, the finite-time mean Cd depends on the initial phase and on the transient after control is switched on. No error bars, no repeated runs, and no convergence check for the mean are provided. The headline claim therefore rests on a single point estimate. Please add multiple controlled runs from different initial conditions (or randomized phases) and report run-to-run variability, or at least show a time-resolved moving average with a clear window and a stationarity check.
  2. [§3.3, Fig. 11; §2.4.2] No comparison is made against a simpler open-loop forcing baseline. The paper states that the control action settles into a quasi-periodic pattern at a frequency that counteracts vortex shedding (§3.3). This raises the possibility that the observed drag reduction is caused by periodic actuation at the shedding frequency, not by the model predictive control or the learned latent dynamics. An open-loop harmonic forcing experiment with the same amplitude and frequency (e.g., sinusoidal blowing/suction at f_sh) is needed to isolate the contribution of the MPC framework. This is load-bearing for the paper's central claim that the framework is effective.
  3. [§3.3, Fig. 12 and surrounding text] The paper acknowledges that, as the wake stabilizes, the model 'departs further from its interpolation region.' Since the MPC cost (Eqs. 13–15) is minimized over model-predicted forces, a biased prediction in that extrapolated regime can lead to suboptimal control actions. Fig. 12 reports the force-prediction error along the horizon, but it does not characterize whether the closed-loop trajectory actually moves outside the training-data distribution. The claim that MPC 'operates effectively despite limitations of the underlying model' would be directly supported by quantifying the closed-loop latent-state distribution relative to the training data (e.g., density overlap or distance from the training manifold), or by a robustness test under model mismatch or perturbed initial conditions. This is a specific, addressable gap.
minor comments (7)
  1. [Eq. (13)] The term labeled 'Mean drag increment' is simply the mean predicted Cd minus the constant Cd,ref. Since Cd,ref is a constant, minimizing this term is equivalent to minimizing the mean Cd; the label is somewhat misleading and could be clarified.
  2. [§2.4.2] The number of optimization iterations is given as 'typically 5' with a learning rate of 10^-3. Please state the exact value used in the reported closed-loop run and whether the result is sensitive to this choice.
  3. [Fig. 11] The colored prediction points in Fig. 11 are visually prominent but the caption does not explain the color mapping or how the 'predicted' values relate to the 'actual' traces. A more explicit caption or a separate legend would improve readability.
  4. [References] Reference [41] is malformed: the author list appears as 'K. D. B. J. Adam, et al.' This should be corrected to Kingma, D.P. & Ba, J. (ICLR 2015).
  5. [Fig. 10] The horizontal axis is labeled 'Number of sensors (Log scale)' but the base of the log scale is not specified. Using explicit tick labels (1, 2, 4, 8, ...) would be clearer.
  6. [§2.1] There is a grammatical issue in 'using the Gym-preCICE [33] wrapper for the [34] coupling library.' Also, all software and acronyms (e.g., Gym-preCICE, preCICE) should be defined at first use.
  7. [Data availability] The data availability statement says materials 'will be made openly available in public repositories upon publication.' Please add repository URLs or a review-access link if possible, and specify the exact datasets/code artifacts that will be released.

Circularity Check

0 steps flagged · score 1.0 of 10

No substantive circularity: the headline drag reduction is measured in the DNS plant, and the model is validated on an independent chirp dataset.

full rationale

The central claim (12.8% drag reduction) is an outcome of the closed-loop DNS plant (Fig. 11), not a quantity reconstructed from the trained model; MPC minimizes a cost defined on model-predicted Cd, but the reported reduction is computed from actual OpenFOAM force coefficients. The predictive model is trained on open-loop frequency-modulated data and tested on a separate chirp signal (Sec. 2.1, 3.1), with sensor selection evaluated against that held-out test set (Fig. 10). The authors explicitly acknowledge that closed-loop operation drives the model partly outside its interpolation region (Sec. 3.3), which is a robustness/extrapolation limitation rather than a circular step. Self-citations (e.g., Refs. [17,21,23]) are background and are not used to justify the drag-reduction claim. The main weakness is the single uncontrolled DNS run, which is an evidence-quality issue, not definitional circularity.

Assumptions & free parameters 8 free parameters · 5 assumptions · 0 invented entities

The framework rests on a large number of hand-chosen hyperparameters—most importantly the MPC cost weights and the sensor count—and on the domain assumption that the 2D Re=500 DNS and the offline-trained model transfer to closed-loop conditions. The empirical 12.8% remains a single unperturbed run.

free parameters (8)
  • MPC cost weights (w_amp, w_Cl, w_smooth) = w_amp=0.1, w_Cl=0.01, w_smooth=8.0
    Chosen by empirical tuning to produce stable control; the reported 12.8% depends on these values.
  • Latent dimension Nz = 8
    Selected by increasing until validation accuracy plateaued; affects the predictive model capacity.
  • Prediction horizon H = 25
    Approximately one shedding cycle; a design choice for the MPC.
  • Optimization iterations per MPC step = 5 with Adam lr=1e-3
    A computational compromise; the closed-loop result may depend on it.
  • Number of selected sensors = 4
    Chosen from the SHAP-importance trade-off curve (Fig. 10) by diminishing returns; a design decision.
  • Data-generation excitation parameters (k_fm, frequency range) = k_fm=0.25, f_base=0.2, AM depth 45%
    Chosen to cover the shedding frequency band; the training distribution of the model.
  • Loss weighting hyperparameters = λ_force=0.25, λ_var=0.02, λ_cov=0.02
    Set by hand for the VICReg and force objectives.
  • History length L and unroll steps k = L=32, k=5
    Model architecture choices affecting temporal memory and training stability.
assumptions (5)
  • domain assumption A 2D simulation at Re=500 captures the essential wake dynamics of a simplified truck configuration relevant for control.
    The entire framework is trained and evaluated in this low-Reynolds 2D setting; conclusions about 'realisable' AFC for trucks assume transferability to 3D/high-Re.
  • domain assumption Surface-mounted pressure sensors provide sufficient information to infer the latent wake state.
    The encoder is trained to map 90 surface pressures to a latent state; the paper demonstrates this in-simulation but does not prove it generally.
  • domain assumption A model trained on one open-loop FM/AM forcing sequence generalizes to closed-loop control actions.
    Stated explicitly in the conclusions; the paper acknowledges the closed-loop trajectory leaves the interpolation region.
  • standard math SHAP GradientExplainer with 100 background and 500 explanation samples yields a reliable sensor ranking.
    An approximation of Shapley values; the paper uses it as the basis for sensor pruning.
  • domain assumption LSTM/MLP training with the described losses converges to a latent space that is predictable and force-informative.
    Standard neural network assumption; no formal convergence guarantee.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A framework for realisable data-driven active flow control using model predictive control applied to a simplified truck wake." pith.science (2026). https://pith.science/paper/QHUBWMBU

@misc{pith2026251011600,
  author       = {Pith},
  title        = {Pith review of: A framework for realisable data-driven active flow control using model predictive control applied to a simplified truck wake},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QHUBWMBU}},
  note         = {Machine review of arXiv:2510.11600}
}
read the original abstract

We present a data-driven active flow control framework designed for deployment with few non-intrusive sensors. The method builds upon Artificial Intelligence driven reduced-order predictive models based on Long-Short-Term Memory (LSTM) networks and efficient gradient-based Model Predictive Control (MPC). The model uses only surface-mounted pressure probes to infer the wake state, and is trained entirely offline on a dataset built with open-loop actuations, thus avoiding the complexities of online learning. Sparsification of the sensors needed for control from an initially large set is achieved using SHapley Additive exPlanations (SHAP). A parsimonious set of sensors is then deployed in closed-loop control with MPC. The framework is tested in numerical simulations of a two-dimensional truck model at Reynolds number 500, with pulsed-jet actuators placed in the rear of the truck to control the wake. The resulting LSTM-MPC achieved a drag reduction of 12.8\%.

Figures

Figures reproduced from arXiv: 2510.11600 by the authors.

Figure 1
Figure 1. Schematic of the flow configuration. Sensor locations depicted in blue and zero-net-mass flow jets [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. DNS mesh and detail of the flow near the rear of the truck, with the jets actuating at maximum [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Sample portion of the modulated control signal applied during training data generation. Top: [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (12 more)
Figure 4
Figure 4. Figure 4: Schematic of the model architecture. A history of sensor observations [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Comparison between the aerodynamic force coefficients estimated by the force decoder and the [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Comparison between the aerodynamic force coefficients estimated by the force decoder and the [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: Pairs plot of latent space variables from train data. Diagonal: histogram of each latent variable. [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: Sample prediction of latent-space variables from test dataset over two shedding cycles. The dashed [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]
Figure 9
Figure 9. Figure 9: Results of the sensor selection algorithm. (a) Sensor importance SHAP values obtained for each [PITH_FULL_IMAGE:figures/full_fig_p017_9.png]
Figure 10
Figure 10. Figure 10: Relative force prediction error (a) and R2 metrics (b) on the test dataset as a function of the number of sensors used. The error is the L1 norm for Cd and Cl , normalized by their respective standard deviations (σ) on the test set. Sensors are added according to thei…
Figure 11
Figure 11. Figure 11: Time series from the closed-loop MPC implementation. Coloured points show the predictions of [PITH_FULL_IMAGE:figures/full_fig_p019_11.png]
Figure 12
Figure 12. Figure 12: L1 error between the aerodynamic forces estimated by the model during MPC control using four sensors and the final actual values. Solid line represents mean error and shaded area represents the standard deviation of the error. Time increment between steps is ∆t = 0.2t…
Figure 13
Figure 13. Figure 13: Kernel density estimate (KDE) plot of the mean pressure coefficient ( [PITH_FULL_IMAGE:figures/full_fig_p020_13.png]
Figure 14
Figure 14. Figure 14: Comparison of the time-averaged streamwise velocity ( [PITH_FULL_IMAGE:figures/full_fig_p021_14.png]
Figure 15
Figure 15. Figure 15: KDE plot of transverse velocity (Uy) fluctuations at a probe in the near wake (10, 0), comparing the uncontrolled (black) and MPC-controlled (blue) cases. contemporary studies on data-driven flow control use velocity or pressure probes placed directly in the wake to o…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. The balance between compactness and forecast accuracy of data-driven latent-space reduced-order models in controlled wake flows

    physics.flu-dyn 2026-07 conditional novelty 5.0 of 10

    In actuated 2D wakes, POD latent dynamics forecast more reliably over long horizons than more compact CAE/VAE latents, revealing a compactness–predictability trade-off for control-oriented ROMs.

Reference graph

Works this paper leans on

47 extracted references · 14 canonical work pages · cited by 1 Pith paper

  1. [1]

    Greenblatt, I

    D. Greenblatt, I. J. Wygnanski, The control of flow separation by periodic exci- tation, Progress in Aerospace Sciences 36 (7) (2000) 487–545. doi:10.1016/S0376- 0421(00)00008-7

    doi:10.1016/s0376- 2000

  2. [2]

    Glezer, M

    A. Glezer, M. Amitay, Synthetic jets, Annual Review of Fluid Mechanics 34 (Volume 34, 2002) (2002) 503–529. doi:10.1146/annurev.fluid.34.090501.094913

    arXiv 2002

  3. [3]

    L. N. Cattafesta, M. Sheplak, Actuators for active flow control, Annual Review of Fluid Mechanics 43 (Volume 43, 2011) (2011) 247–272. doi:10.1146/annurev-fluid-122109- 160634

    work page doi:10.1146/annurev-fluid-122109- 2011

  4. [4]

    Littlewood, M

    R. Littlewood, M. A. Passmore, Aerodynamic drag reduction of a simplified square- back vehicle using steady blowing, Experiments in fluids 53 (2) (2012) 519–529. doi:10.1007/s00348-012-1306-4

    work page doi:10.1007/s00348-012-1306-4 2012

  5. [5]

    McNally, E

    J. McNally, E. Fernandez, G. Robertson, R. Kumar, K. Taira, F. Alvi, Y. Yamaguchi, K. Murayama, Drag reduction on a flat-back ground vehicle with active flow con- trol, Journal of Wind Engineering and Industrial Aerodynamics 145 (2015) 292–303. doi:10.1016/j.jweia.2015.03.006

    work page doi:10.1016/j.jweia.2015.03.006 2015

  6. [6]

    J. J. Cerutti, C. Sardu, G. Cafiero, G. Iuso, Active flow control on a square-back road vehicle, Fluids 5 (2) (2020) 55. doi:10.3390/fluids5020055

    work page doi:10.3390/fluids5020055 2020

  7. [7]

    Amico, G

    E. Amico, G. Cafiero, G. Iuso, Deep reinforcement learning for active control of a three- dimensional bluff body wake, Physics of Fluids 34 (10) (2022). doi:10.1063/5.0108387

    work page doi:10.1063/5.0108387 2022

  8. [8]

    Amico, J

    E. Amico, J. Serpieri, G. Iuso, G. Cafiero, Flow topology of deep reinforcement learning drag-reduced bluff body wakes, Physics of Fluids 36 (8) (2024). doi:10.1063/5.0217692

    work page doi:10.1063/5.0217692 2024

Show all 47 references
  1. [9]

    S. L. Brunton, B. R. Noack, Closed-loop turbulence control: Progress and challenges, Applied Mechanics Reviews 67 (5) (2015) 050801. doi:10.1115/1.4031175. 24

    2015 doi

  2. [10]

    S. L. Brunton, B. R. Noack, P. Koumoutsakos, Machine learning for fluid me- chanics, Annual Review of Fluid Mechanics 52 (Volume 52, 2020) (2020) 477–508. doi:10.1146/annurev-fluid-010719-060214

    2020 doi

  3. [11]

    Regazzoni, S

    F. Regazzoni, S. Pagani, M. Salvador, L. Dede’, A. Quarteroni, Learning the intrin- sic dynamics of spatio-temporal processes through latent dynamics networks, Nature Communications 15 (1) (2024) 1834. doi:10.1038/s41467-024-45323-x

    2024 doi

  4. [12]

    E. F. Camacho, C. Bordons, Model Predictive Control, 2nd Edition, Advanced Textbooks in Control and Signal Processing, Springer London, London, 2013. doi:10.1007/978-0-85729-398-5

    2013 doi

  5. [13]

    J. H. Lee, Model predictive control: Review of the three decades of development, International Journal of Control, Automation and Systems 9 (3) (2011) 415–424. doi:10.1007/s12555-011-0300-6

    2011 doi

  6. [14]

    T. R. Bewley, P. Moin, R. Temam, DNS-based predictive control of turbulence: an optimal benchmark for feedback algorithms, Journal of Fluid Mechanics 447 (2001) 179–225. doi:10.1017/S0022112001005821

    2001 doi

  7. [15]

    Morton, A

    J. Morton, A. Jameson, M. J. Kochenderfer, F. Witherden, Deep dynamical modeling and control of unsteady fluid flows, Advances in Neural Information Processing Systems 31 (2018)

    2018

  8. [16]

    Bieker, S

    K. Bieker, S. Peitz, S. L. Brunton, J. N. Kutz, M. Dellnitz, Deep model predictive flow control with limited sensor data and online learning, Theoretical and computational fluid dynamics 34 (4) (2020) 577–591. doi:10.1007/s00162-020-00520-4

    2020 doi

  9. [17]

    Marra, A

    L. Marra, A. Meilán-Vila, S. Discetti, Self-tuning model predictive control for wake flows, Journal of Fluid Mechanics 983 (2024). doi:10.1017/jfm.2024.47

    2024 doi

  10. [18]

    S. L. Brunton, J. L. Proctor, J. N. Kutz, Discovering governing equations from data by sparse identification of nonlinear dynamical systems, Proceedings of the national academy of sciences 113 (15) (2016) 3932–3937. doi:10.1073/pnas.1517384113

    2016 doi

  11. [19]

    Kaiser, J

    E. Kaiser, J. N. Kutz, S. L. Brunton, Sparse identification of nonlinear dynamics for model predictive control in the low-data limit, Proceedings of the Royal Society A 474 (2219) (2018) 20180335. doi:10.1098/rspa.2018.0335

    2018

  12. [20]

    Sobal, W

    V. Sobal, W. Zhang, K. Cho, R. Balestriero, T. G. J. Rudner, Y. LeCun, Learning from reward-free offline data: A case for planning with latent dynamics models, arXiv preprint arXiv:2502.14819 (2025)

    2025

  13. [21]

    Solera-Rico, C

    A. Solera-Rico, C. Sanmiguel Vila, M. Gómez-López, Y. Wang, A. Almash- jary, S. T. Dawson, R. Vinuesa,β-variational autoencoders and transformers for reduced-order modelling of fluid flows, Nature Communications 15 (1) (2024) 1361. doi:10.1038/s41467-024-45578-4. 25

    2024 doi

  14. [22]

    Z. Liu, D. Beckers, J. D. Eldredge, Model-based reinforcement learning for con- trol of strongly disturbed unsteady aerodynamic flows, AIAA Journal (2025). doi:10.2514/1.J064790

    2025 doi

  15. [23]

    Fukagata, K

    K. Fukagata, K. Fukami, Compressing fluid flows with nonlinear machine learning: mode decomposition, latent modeling, and flow control, Fluid Dynamics Research (2025). doi:10.1088/1873-7005/ade8a2

    2025 doi

  16. [24]

    Fukami, K

    K. Fukami, K. Taira, Grasping extreme aerodynamics on a low-dimensional manifold, Nature Communications 14 (1) (2023) 6480

    2023

  17. [25]

    Fukami, K

    K. Fukami, K. Taira, Observable-augmented manifold learning for multi-source turbu- lent flow data, Journal of Fluid Mechanics 1010 (2025) R4. doi:10.1017/jfm.2025.383

    2025 doi

  18. [26]

    Rabault, M

    J. Rabault, M. Kuchta, A. Jensen, U. Réglade, N. Cerardi, Artificial neural networks trained through deep reinforcement learning discover control strategies for active flow control, Journal of Fluid Mechanics 865 (2019) 281–302. doi:10.1017/jfm.2019.62

    2019 doi

  19. [27]

    Castellanos, G

    R. Castellanos, G. Y. Cornejo Maceda, I. de la Fuente, B. R. Noack, A. Ianiro, S. Dis- cetti, Machine-learning flow control with few sensor feedback and measurement noise, Physics of Fluids 34 (4) (2022) 047118. doi:10.1063/5.0087208

    2022 doi

  20. [28]

    T. C. Déda, W. R. Wolf, S. T. M. Dawson, Neural networks in feed- back for flow analysis and control, Physical Review Fluids 9 (2024) 063904. doi:10.1103/PhysRevFluids.9.063904

    2024 doi

  21. [29]

    J. B. Rawlings, D. Q. Mayne, M. M. Diehl, Model Predictive Control: Theory, Com- putation, and Design, 2nd Edition, Nob Hill Publishing, 2017

    2017

  22. [30]

    S. M. Lundberg, S.-I. Lee, A unified approach to interpreting model predictions, Ad- vances in neural information processing systems 30 (2017)

    2017

  23. [31]

    B. Amos, I. Jimenez, J. Sacks, B. Boots, J. Z. Kolter, Differentiable mpc for end-to-end planning and control, Advances in neural information processing systems 31 (2018)

    2018

  24. [32]

    B. L. Storms, J. C. Ross, J. T. Heineck, S. M. Walker, D. M. Driver, G. G. Zilliac, D. P. Bencze, An experimental study of the ground transportation system (gts) model in the nasa ames 7 by 10 ft wind tunnel, Tech. Rep. NASA/TM-2001-209621, NASA Ames Research Center (February 2001)

    2001

  25. [33]

    Shams, A

    M. Shams, A. H. Elsheikh, Gym-precice: Reinforcement learning environments for ac- tive flow control, SoftwareX 23 (2023) 101446. doi:10.1016/j.softx.2023.101446

    2023

  26. [34]

    Chourdakis, K

    G. Chourdakis, K. Davis, B. Rodenberg, M. Schulte, F. Simonis, B. Ueker- mann, G. Abrams, H. Bungartz, L. Cheung Yau, I. Desai, K. Eder, R. Hertrich, F. Lindner, A. Rusch, D. Sashko, D. Schneider, A. Totounferoush, D. Volland, 26 P. Vollmer, O. Koseomur, preCICE v2: A sustaina...

    2022 doi

  27. [35]

    Chourdakis, D

    G. Chourdakis, D. Schneider, B. Uekermann, OpenFOAM-preCICE: Coupling Open- FOAM with external solvers for multi-physics simulations, OpenFOAM®Journal 3 (2023) 1–25. doi:10.51560/ofj.v3.88

    2023 doi

  28. [36]

    Barros, J

    D. Barros, J. Borée, B. R. Noack, A. Spohn, T. Ruiz, Bluff body drag manipulation using pulsed jets and coanda effect, Journal of Fluid Mechanics 805 (2016) 422–459. doi:10.1017/jfm.2016.508

    2016 doi

  29. [37]

    Hochreiter, J

    S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Computation 9 (8) (1997) 1735–1780. doi:10.1162/neco.1997.9.8.1735

    1997 doi

  30. [38]

    P. C. Blaud, P. Chevrel, F. Claveau, P. Haurant, A. Mouraud, ResNet and PolyNet based identification and (MPC) control of dynamical systems: a promising way, IEEE Access 11 (2023) 20657–20672. doi:10.1109/ACCESS.2022.3196920

    2023

  31. [39]

    K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in: Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778

    2016

  32. [40]

    Bardes, J

    A. Bardes, J. Ponce, Y. LeCun, Vicreg: Variance-invariance-covariance regularization for self-supervised learning (2022). arXiv:2105.04906

    2022 arXiv

  33. [41]

    K. D. B. J. Adam, et al., A method for stochastic optimization, arXiv preprint arXiv:1412.6980 1412 (6) (2014)

    2014 arXiv

  34. [42]

    L. S. Shapley, A Value for n-Person Games, Princeton University Press, Princeton, 1953, Ch. 17, pp. 307–318. doi:10.1515/9781400881970-018

    1953 doi

  35. [43]

    Sundararajan, A

    M. Sundararajan, A. Taly, Q. Yan, Axiomatic attribution for deep networks, in: Inter- national conference on machine learning, PMLR, 2017, pp. 3319–3328

    2017

  36. [44]

    Paris, S

    R. Paris, S. Beneddine, J. Dandois, Robust flow control and optimal sensor place- ment using deep reinforcement learning, Journal of Fluid Mechanics 913 (2021) A25. doi:10.1017/jfm.2020.1170

    2021

  37. [45]

    Paszke, S

    A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, et al., Pytorch: An imperative style, high-performance deep learning library, Advances in neural information processing systems 32 (2019)

    2019

  38. [46]

    Pastoor, L

    M. Pastoor, L. Henning, B. R. Noack, R. King, G. Tadmor, Feedback shear layer control for bluff body drag reduction, Journal of fluid mechanics 608 (2008) 161–196. doi:10.1017/S0022112008002073. 27

    2008 doi

  39. [47]

    Dalla Longa, A

    L. Dalla Longa, A. Morgans, J. Dahan, Reducing the pressure drag of a d-shaped bluff body using linear feedback control, Theoretical and Computational Fluid Dynamics 31 (5) (2017) 567–577. doi:10.1007/s00162-017-0420-6. 28

    2017 doi

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.